The most studied and most replicated ergogenic supplement of the last three decades. Cheap, safe, made of three amino acids, and effective for the specific thing it does — buffering ATP regeneration during short, high-intensity work. The newer evidence on cognition, bone, and mood is what makes creatine interesting beyond the gym.
Creatine is a small molecule synthesized from glycine, arginine, and methionine in the liver and kidneys. Once in tissue, creatine kinase phosphorylates it into phosphocreatine, which serves as a rapidly mobilizable phosphate donor to regenerate ATP from ADP. The system dominates energy supply in the first 10–15 seconds of high-intensity work — the rep range that builds strength and power output.
Supplementing 3–5 g/day for several weeks expands the muscle phosphocreatine pool by ~10–40% above baseline, with the largest increases in people whose baseline is lowest (vegetarians, older adults). That headroom translates into more reps, slightly more weight, and a modest but real increase in lean mass over training cycles [Kreider 2017].
The brain has its own phosphocreatine pool, refreshed by a separate but slower transport across the blood-brain barrier. That separation is why the cognitive benefits of supplementation seem to require larger doses or longer time courses than the muscle effects, and why the people who benefit most are those with constrained substrate — sleep-deprived and vegetarians.
| Phase | Dose | Timing | Note |
|---|---|---|---|
| Standard maintenance | 3–5 g/day | Any time, daily | Plateau in ~4 weeks |
| Loading (optional) | 20 g/day in 4 doses × 5–7 days | Then 3–5 g/day | Faster saturation, more GI side effects |
| Cognitive target | 5–10 g/day | Daily, ongoing | Higher doses studied for brain effects |
| Heavier athletes | 5 g/day per ~70 kg | Daily | Larger muscle mass, larger pool to fill |
HCl, buffered ("Kre-Alkalyn"), ethyl ester, and other novel forms have not outperformed monohydrate in head-to-head trials, and they cost more. Monohydrate is what every major trial used. Buy that.
Timing relative to workouts does not meaningfully matter for the muscle effect — daily dosing matters more than the hour. Co-ingesting with carbs or protein modestly increases uptake via insulin, but the saturation endpoint is the same. Cycling off is not necessary.
Rae and colleagues showed in 2003 that 5 g/day of creatine for six weeks improved working memory (backward digit span) and Raven's Progressive Matrices in young vegetarian adults — both tasks that load processing speed [Rae 2003]. A 2018 systematic review by Avgerinos and colleagues pooled the data and concluded that creatine appears to improve short-term memory and intelligence/reasoning, with effects most consistent under conditions of stress, sleep deprivation, or low dietary intake [Avgerinos 2018].
Translation: an already-replete omnivore eating beef daily may not get a noticeable cognitive bump from supplementation. A vegetarian, a sleep-deprived shift worker, or an older adult with declining endogenous synthesis is much more likely to feel something. This is consistent with the pool-saturation model — you can't fill what's already full.
Bone-density signal is newer. Trials in postmenopausal women combining creatine with resistance training have shown attenuated loss of proximal femur bone mineral density over 12 months [Chilibeck 2015]. The effect is small, real, and worth tracking.
In 2009, van der Merwe and colleagues studied 20 college-aged rugby players given a creatine loading protocol. After seven days of loading, DHT (dihydrotestosterone) rose by roughly 56% from baseline and remained about 33% above baseline through three weeks of maintenance dosing [van der Merwe 2009]. This is the single study every forum thread is referencing.
Two caveats matter. First, the absolute DHT values stayed within the normal physiological range — the percent change was eye-catching, the biology less so. Second, the study did not measure hair loss, hair density, follicle count, or anything else hair-related. It measured a hormone. Subsequent literature looking at creatine's effects on androgens has not consistently replicated the magnitude of the DHT change, and no controlled study to date has demonstrated creatine causing hair loss. Someone with strong familial androgenic alopecia who notices accelerated shedding on creatine is reasonable to discontinue; a population-level concern is not supported.
| Form | Typical price | Cost per day (5 g) | Note |
|---|---|---|---|
| Monohydrate (bulk) | $15–25 / kg | ~$0.10 | Buy Creapure if you want a guaranteed source |
| Capsules | $25–40 / month | $0.80–$1.30 | Convenience tax |
| HCl / "advanced" | $40–60 / month | $1.30–$2.00 | No clinical advantage |
Creatine is the rare supplement where the evidence is unusually strong, the dose is unusually simple, and the cost is unusually low. If someone trains, is vegetarian, sleeps poorly, or is over 50, the case is straightforward. Monohydrate, 3–5 g/day, indefinitely. See the longevity hub for context and the sleep stack piece for how creatine fits alongside other interventions.